crhea93

crhea93

crhea93

crhea93

crhea93

# Define datasets
X = [1, 1.25, 1.8, 1.9, 2.21, 2.85]
Y = [1.1, 1.14, 1.8, 1.85, 2.25, 2.9]
print(X,Y)
# Let's see what this looks like
with plt.xkcd():
    plt.scatter(X, Y)
    plt.show()

crhea93

# Calculate means and normalize
x_mean = np.mean(X)
X_adj = [x-x_mean for x in X]
y_mean = np.mean(Y)
Y_adj = [y - y_mean for y in Y]
# Caculate Covariance Matrix
cov = np.cov(X_adj, Y_adj)
print(cov)

crhea93

lam, v = np.linalg.eig(cov)
print("Eigenvalues:")
print(lam)
print("Eigenvectors:")
print(v)
# Sort
eigvals, eigvects = (list(t) for t in zip(*sorted(zip(lam, v))))

# And now we can see what this looks like!
x_s = np.linspace(np.min(X_adj), np.max(X_adj), 10)

crhea93

import corner
import emcee
from google.colab import drive
import pandas as pd
import matplotlib.pyplot as plt
import mplcyberpunk
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
plt.style.use("cyberpunk")

crhea93

data = pd.read_csv(DATASET_PATH, delimiter=' ')
data['M_BH'] = np.log10(data['M_BH'])
data['M_low'] = np.log10(data['M_low'])
data['M_high'] = np.log10(data['M_high'])
# Change low and high to errors
data['M_low'] = data['M_BH'] - data['M_low']
data['M_high'] = data['M_high'] - data['M_BH']

crhea93

n = 50  # Size of spectrum
N = 5000  # Number of spectra
def gaussian(x, mu, sig):
    return 1*np.exp(-np.power(x - mu, 2.) / (2 * np.power(sig, 2.)))

def conv_mat(n):
    """
    Create convolution matrix that is an identity matrix with noise
    """
    conv_mat = np.eye(n)+np.random.normal(0, 0.05, (n,n))